Clinical utility of early improvement to predict response or remission in acute mania: focus on olanzapine and risperidone.

OBJECTIVE

To evaluate early improvement associated with atypical antipsychotic treatment as a predictor of later response or remission among patients experiencing an acute manic or mixed episode without psychotic features.

METHOD: A post hoc analysis was performed on data from a 3-week, randomized, double-blind clinical trial of olanzapine (N = 147) or risperidone (N = 127) to treat inpatients aged 18-70 years meeting DSM-IV criteria for bipolar I disorder. Early improvement, measured as percent change (≥ 25% and ≥ 50% cut points) in the Young Mania Rating Scale (YMRS) total score, was assessed after 2 days and 1 week of treatment. Receiver operating characteristic curves, sensitivity and specificity, and positive and negative predictive values were calculated to determine whether early improvement predicted endpoint (week 3) response or remission. The study was conducted from July 2001 through June 2002.

RESULTS

Among 234 patients with ≥ 25% reduction in YMRS total score at week one, 167 (71.4%) responded and 121 (51.7%) remitted at endpoint. Of the 40 patients with < 25% improvement, 25% (n = 10) responded and 5% (n = 2) remitted at endpoint. A total of 157 patients had a ≥ 50% reduction in week 1 YMRS total score, of whom 132 (84.1%) responded and 101 (64.3%) remitted at endpoint. Of the 117 patients with < 50% improvement, 45 (38.5%) responded and 22 (18.8%) remitted at endpoint.

CONCLUSIONS

Improvement in manic or mixed symptoms at week 1 appears to be a good predictor of treatment outcome. Patients not having sufficient improvement (< 25% reduction in YMRS score) were less likely to reach response or remission by week 3. Patients who achieved response by week 1 (≥ 50% reduction in YMRS score) were likely to remain responders at endpoint. These data suggest the potential to assess benefit in the treatment of manic or mixed symptoms within 1 week of initiating olanzapine or risperidone.